A knowledge-based system for individual income and transfer tax planning

Recent developments in computer science include the creation of systems that are capable of solving difficult applications problems requiring expert knowledge for their solution. This type of system had not yet been developed for solving income and transfer (estate and gift) tax problems. The purpose of the dissertation was to develop such a system for solving problems concerning income and transfer tax planning for individuals so that the feasibility of using an artificial intelligence (AI) model for tax planning could be determined. In order to develop such a system, a theoretical structure and a set of decision rules were specified and programmed using a rule-based system that had already been used for medical diagnosis. Once the system was developed, its problem-solving capabilities were refined and verified by a panel of tax experts. The dissertation is organized in the following manner. In the first chapter, AI is defined and discussed; the apparent convergence of AI, cognitive psychology, and decision theory is explored; and the implications of AI for tax planning models are described. In the next two chapters, the AI system that was utilized and the tax planning decision that was modeled are described. Refinement and verification of the model and results of the study are discussed in chapters four and five. The final chapter includes an interpretation of results and suggested extensions of the research. With respect to Federal tax law, the dissertation is up-to-date through July 31, 1981.